| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Wed, 19 Apr 2017 07:12:40 +0800
From: Baoquan He <bhe@...hat.com>
To: Kees Cook <keescook@...omium.org>
Cc: LKML <linux-kernel@...r.kernel.org>,
Dave Jiang <dave.jiang@...el.com>,
Dan Williams <dan.j.williams@...el.com>,
"H. Peter Anvin" <hpa@...or.com>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...nel.org>, Dave Young <dyoung@...hat.com>,
Ingo Molnar <mingo@...hat.com>,
"x86@...nel.org" <x86@...nel.org>, Yinghai Lu <yinghai@...nel.org>,
Borislav Petkov <bp@...e.de>
Subject: Re: [PATCH 3/4] KASLR: Handle memory limit specified by memmap and
mem option
On 04/18/17 at 01:36pm, Kees Cook wrote:
> On Mon, Apr 17, 2017 at 6:34 AM, Baoquan He <bhe@...hat.com> wrote:
> > Option mem= will limit the max address system can use. Any memory
> > region above the limit will be removed. And memmap=nn[KMG] which
> > has no offset specified has the same behaviour as mem=. KASLR need
> > consider this when choose the random position for decompressing
> > kernel. Do it in this patch.
> >
> > Signed-off-by: Baoquan He <bhe@...hat.com>
> > Cc: "H. Peter Anvin" <hpa@...or.com>
> > Cc: Thomas Gleixner <tglx@...utronix.de>
> > Cc: Ingo Molnar <mingo@...hat.com>
> > Cc: x86@...nel.org
> > Cc: Kees Cook <keescook@...omium.org>
> > Cc: Yinghai Lu <yinghai@...nel.org>
> > Cc: Borislav Petkov <bp@...e.de>
> > ---
> > arch/x86/boot/compressed/kaslr.c | 53 +++++++++++++++++++++++++++++++---------
> > 1 file changed, 42 insertions(+), 11 deletions(-)
> >
> > diff --git a/arch/x86/boot/compressed/kaslr.c b/arch/x86/boot/compressed/kaslr.c
> > index 36ab429..5361abd 100644
> > --- a/arch/x86/boot/compressed/kaslr.c
> > +++ b/arch/x86/boot/compressed/kaslr.c
> > @@ -67,6 +67,10 @@ int mem_avoid_memmap_index;
> > extern unsigned long get_cmd_line_ptr(void);
> >
> >
> > +/* Store memory limit specified by "mem=nn[KMG]" or "memmap=nn[KMG]" */
> > +unsigned long long mem_limit = ULLONG_MAX;
>
> I would either make this 0 or ULLONG_MAX - 1 (see below).
I prefer "ULLONG_MAX-1". Please see reason below
>
> > +
> > +
> > enum mem_avoid_index {
> > MEM_AVOID_ZO_RANGE = 0,
> > MEM_AVOID_INITRD,
> > @@ -117,15 +121,18 @@ parse_memmap(char *p, unsigned long long *start, unsigned long long *size)
> > return -EINVAL;
> >
> > switch (*p) {
> > - case '@':
> > - /* Skip this region, usable */
> > - *size = 0;
> > - *start = 0;
> > case '#':
> > case '$':
> > case '!':
> > *start = memparse(p + 1, &p);
> > return 0;
> > + case '@':
> > + /* Skip this region, usable */
> > + *size = 0;
>
> Now it looks like we're intentionally falling through. A comment
> should be included to indicate it.
Yes, as I replied in patch 2/4, it's intended. Will add comment to
indicate.
>
> > + default:
> > + /* Avoid the region which is above the amount limit */
> > + *start = 0;
> > + return 0;
> > }
> >
> > return -EINVAL;
> > @@ -151,9 +158,14 @@ static void mem_avoid_memmap(char *str)
> > if (rc < 0)
> > break;
> > str = k;
> > - /* A usable region that should not be skipped */
> > - if (size == 0)
> > +
> > + if (start == 0) {
> > + /* Store the specified memory limit if size > 0 */
> > + if (size > 0)
> > + mem_limit = size;
> > +
> > continue;
> > + }
> >
> > mem_avoid[MEM_AVOID_MEMMAP_BEGIN + i].start = start;
> > mem_avoid[MEM_AVOID_MEMMAP_BEGIN + i].size = size;
> > @@ -173,6 +185,7 @@ static int handle_mem_memmap(void)
> > char tmp_cmdline[COMMAND_LINE_SIZE];
> > size_t len = strlen((char *)args);
> > char *param, *val;
> > + u64 mem_size;
> >
> > len = (len >= COMMAND_LINE_SIZE) ? COMMAND_LINE_SIZE - 1 : len;
> > memcpy(tmp_cmdline, args, len);
> > @@ -195,8 +208,18 @@ static int handle_mem_memmap(void)
> > return -1;
> > }
> >
> > - if (!strcmp(param, "memmap"))
> > + if (!strcmp(param, "memmap")) {
> > mem_avoid_memmap(val);
> > + } else if (!strcmp(param, "mem")) {
> > + char *p = val;
> > +
> > + if (!strcmp(p, "nopentium"))
> > + continue;
> > + mem_size = memparse(p, &p);
> > + if (mem_size == 0)
> > + return -EINVAL;
> > + mem_limit = mem_size;
> > + }
> > }
> >
> > return 0;
> > @@ -432,7 +455,8 @@ static void process_e820_entry(struct e820entry *entry,
> > {
> > struct mem_vector region, overlap;
> > struct slot_area slot_area;
> > - unsigned long start_orig;
> > + unsigned long start_orig, end;
> > + struct e820entry cur_entry;
> >
> > /* Skip non-RAM entries. */
> > if (entry->type != E820_RAM)
> > @@ -446,8 +470,15 @@ static void process_e820_entry(struct e820entry *entry,
> > if (entry->addr + entry->size < minimum)
> > return;
> >
> > - region.start = entry->addr;
> > - region.size = entry->size;
> > + /* Ignore entries above memory limit */
> > + end = min(entry->size + entry->addr - 1, mem_limit);
> > + if (entry->addr >= end)
> > + return;
> > + cur_entry.addr = entry->addr;
> > + cur_entry.size = end - entry->addr + 1;
> > +
> > + region.start = cur_entry.addr;
> > + region.size = cur_entry.size;
>
> I find the manipulation of entry->addr +/- 1 confusing; it should just
> be mem_limit that is adjusted:
>
> end = min(entry->size + entry->addr, mem_limit + 1);
The first one is preferred. Since if no memory limit specifed, we still have
a limit, 64T-1 or UULONG_MAX -1, in logic more understandable. And in
code lines it's less.
Will change with your suggestion.
Thanks a log for your comments!
>
> And maybe to avoid mem_limit being giant by default, maybe have "0" be special?
>
> cur_entry.addr = entry->addr;
> if (mem_limit) {
> unsigned long end = min(entry->size + entry->addr, mem_limit + 1);
> if (entry->addr > end)
> return;
> cur_entry.size = end - entry->addr;
> } else {
> cur_entry.size = entry->size;
> }
>
> or something... and maybe move the whole thing earlier so other tests
> that examine entry->size are checked with the new adjusted value.
>
> -Kees
>
> >
> > /* Give up if slot area array is full. */
> > while (slot_area_index < MAX_SLOT_AREA) {
> > @@ -461,7 +492,7 @@ static void process_e820_entry(struct e820entry *entry,
> > region.start = ALIGN(region.start, CONFIG_PHYSICAL_ALIGN);
> >
> > /* Did we raise the address above this e820 region? */
> > - if (region.start > entry->addr + entry->size)
> > + if (region.start > cur_entry.addr + cur_entry.size)
> > return;
> >
> > /* Reduce size by any delta from the original address. */
> > --
> > 2.5.5
> >
>
>
>
> --
> Kees Cook
> Pixel Security
Powered by blists - more mailing lists